You are on page 1of 8

See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/26445517

Evaluation of Slope Assessment Systems for


Predicting Landslides of Cut Slopes in Granitic
and Meta-sediment Formations

Article in American journal of environmental sciences April 2006


DOI: 10.3844/ajessp.2006.135.141 Source: DOAJ

CITATIONS READS

16 17

3 authors:

Suhaimi Jamaludin Bujang B. K. Huat


Slope Engineering Branch Universiti Putra Malaysia
24 PUBLICATIONS 48 CITATIONS 230 PUBLICATIONS 1,311 CITATIONS

SEE PROFILE SEE PROFILE

Husaini Omar
Universiti Putra Malaysia
59 PUBLICATIONS 153 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Long-term settlement prediction of open dumping area using Monte Carlo Simulation method View
project

Acid Rain Intrusion Effects on Slope Failure Phenomena and Mechanisms View project

All content following this page was uploaded by Bujang B. K. Huat on 30 July 2016.

The user has requested enhancement of the downloaded file.


American Journal of Environmental Sciences 2 (4): 135-141, 2006
ISSN 1553-345X
2006 Science Publications

Evaluation of Slope Assessment Systems for Predicting Landslides of Cut Slopes in


Granitic and Meta-sediment Formations
1
Suhaimi Jamaludin, 2Bujang B.K. Huat and 2Husaini Omar
1
Public Works Department, Malaysia
2
Faculty of Engineering, University Putra Malaysia

Abstract: In Malaysia, slope assessment systems (SAS) are widely used in assessing the instability of
slope or the probability of occurrence and the likely severity of landslides. These SAS can be derived
based on either one particular approach or combination of several approaches of landslide assessments
and prediction. This study overviews four slope assessment systems (SAS) developed in Malaysia for
predicting landslide at a large-scale assessments. They are the Slope Maintenance System (SMS),
Slope Priority Ranking System (SPRS), Slope Information Management System (SIMS) and the Slope
Management and Risk Tracking System (SMART). An attempt is made to evaluate the accuracy of the
SAS in predicting landslides based on slope inventory data from 139 cut slopes in granitic formation
and 47 cut slopes in meta-sediment formation, which are the two most common rock/soil formations
found in Malaysia. Based on this study, it was found that none of existing SAS is satisfactory in
predicting landslides of cut slopes in granitic formation, for various reasons such as the use of hazard
score developed from another country, insufficient data base, oversimplified approach and use of data
base derived from different rock/soil formations. However for the case of cut slope in meta-sediment,
the Slope Management and Risk Tracking System (SMART) was found to be satisfactory with 90%
prediction accuracy. The current database of SMART is largely based on meta-sediment formation.

Key words: Landslides, cut slopes, tropical soils, slope assessment systems, granitic formation, meta-
sediment formation

INTRODUCTION units[3]. Flows consist of movement of slurry of soil and


loose rocks down slope in a manner analogous to a
Landslides have caused large numbers of casualties viscous fluid. Falls are incidence of masses of rocks
and huge economic losses in hilly and mountainous detaching from a steep slope and descending by free
areas of the world. In tropical countries where annual fall, rolling or bouncing.
rainfall can reach as high as 4500 mm and high Landslide assessment for the purpose of estimating
temperatures around the year, cause intense weathering the probability of occurrence and likely severity of
and formation of thick soil and weathered rock profile. landslides can be carry out by various methods, namely
With these set of climate and geological condition, the statistical method, landslide inventory method,
combined with other causative factors, landslide is one heuristic approach and deterministic approach[4].
of the most destructive natural disasters in tropical Tangestani[5] describe attempt to use of fuzzy set theory
region. Malaysia is one of the countries located in the analysis, while Yi et al.[6] use fractal dimension, a
tropical region. During the period from 1993 to 2006, a
mathematical theory that describes the quality of
number of major landslides were reported in Malaysia,
complex shapes of images in the nature, in evaluating
involving fill and cut of natural slopes, which results in
death of people. landslide hazard.
The most common type of landslides in Malaysia is In Malaysia, there are at least eight slope
assessment systems (SAS) that have been developed
shallow slide where the slide surface is usually less than
over the last ten years. Four of these SAS, all meant for
4 m deep and occurs during or immediately after
large-scale assessment, namely the Slope Maintenance
intense rainfall[1]. These slides commonly occur in System (SMS), Slope Priority Ranking System (SPRS),
residual soils mantles of grade V and grade VI Slope Information Management System (SIMS) and the
according to the commonly used classification systems Slope Management and Risk Tracking System
of Little[2]. Other types of landslides found are deep- (SMART), all developed by the Public Works
seated slides, debris flow and geologically controlled Department (PWD) of Malaysia[7-10] are described in
failures such as wedge failures and rock fall. Slide is this study. Large-scale assessment refers to use of maps
defined as downward displacement or soil (or rock) of scale between 1:5,000 and 1:15,000. Despite the
sliding along one or more failure surfaces, rotational for enormous effort given to develop the slope assessment
the case of few units; translational for the case of many systems, no attempt has been made to date to validate

Corresponding Author: Husaini Omar, Faculty of Engineering, University Putra Malaysia


135
Am. J. Environ. Sci., 2 (4): 135-141, 2006

the accuracy of any of these SAS in predicting the Table 1: Hazard weighting for cut slopes of main range granite
likelihood of landslides (slope failures). The accuracy used in the SMS [7]
Parameter Sub-parameter Weighting
or reliability in predicting future landslides is crucial to
Age in years < 8, 8 -11 & >11 0 to 2.0
any SAS. Incorrect prediction will expose lives and Culverts Culvert & No Culvert 0 to 2.0
economy to danger or hazard if a slope or an area that Erosion No erosion, Sheet, Rill 0 to 2.0
should has a high hazard level is incorrectly & Gully
classified/predicted as with low hazard level. On other Percentage of 0 to 100 percent 0 to 2.0
hand, if a slope or an area that should have a low hazard feature uncovered
Feature aspect in 0 to 360 degrees 0 to 2.0
level is incorrectly predicted as high hazard level, it has degrees
financial implication because money will be spend to Rock condition Claystone, 0 to 2.0
stabilized the stable (not failed) slope. This research profile Conglomerate, Granite,
described a study that has been made to validate the Limestone, Phylite &
existing SAS based on slope inventory data from 139 Sandstone
cut slopes in granitic formation and 47 cut slopes
Table 2: Hazard weighting for cut slopes of meta-sediment use in
underlain by meta-sediment formations. These are the
the SMS[7]
two major rock/soil formations found in Malaysia. Parameter Sub-parameter Weighting
Granite is the major rock that underlies virtually Number of water courses 0 to 2 0 to 2.0
every major mountain range with summits exceeding within features
2,000m in Malaysia. About 30% (5,000km) of major Rock condition profile Granite, Limestone, 0 to 2.0
trunk roads which involve many cut slopes, traverse Phyllite and
Sandstone.
through or located on hilly and mountainous areas of Erosion No Erosion, Sheet, 0 to 2.0
Malaysia. Some 75% of the roads that traversed Rill & Gully.
through the hilly and mountainous areas are cut through Distance to ridge or 0 to >200. 0 to 2.0
and/or underlain by granitic formation. The remaining gully in meters
25% of the roads are cut through or underlain by the Feature aspect in degrees 0 to 360 degrees 0 to 2.0
Slope angle in degrees 0 to 90 degrees 0 to 2.0
meta-sediment formations (mudstone, sandstone and
siltstone). These mountainous roads experienced
Table 3: Hazard level and range of hazard rating in percentage use
numerous numbers of landslides occurrences in the in the SMS[7]
past, usually during the wet (rainy) season from
Hazard Score Hazard Rating / Level
October to January, which had caused disruption to
traffic, injuries and losses of life. A study carried out in 80.1% -100% Very High
the year 2000 along six selected hilly and mountainous 60.1% - 80% High
roads shows that out of 444 landslides of various types 40.1% 60% Medium
(shallow slides, deep seated slides, debris flow and rock 20.1% 40% Low
fall), 420 occur in cut and natural slopes[9]. The other 24 0% 20% Very Low
slides occur in embankment (fill) slopes.
For example out of 100 known landslides, 5
SLOPE ASSESSMENT SYSTEMS
numbers are in the range of 8 to 11 years old slope, so
The Slope Maintenance System (SMS) was the the weighting for this range of age is 0.1 (5 divided by
first slope assessment system developed by the Public 100 and multiply by 2). Using this method, the
Works Department (PWD) of Malaysia, as part of the weightings for other slope parameters were established.
East-West Highway long-term preventive measures[7]. Table 1 shows example of hazard weighting for cut
Statistical method using discriminant analyses based on slopes in granitic formation as used in the SMS. The
slope type (embankment/fill and cut/natural slope) were hazard weighting was developed based on 74 cut slopes
used to determine the hazard values. The parameters (of which 31 was failed slopes) in the main range
captured for each slope include age of the cut slope, granite formation along the East-West Highway of
batter height, bench width, ratio of crest length to edge Peninsular Malaysia.
length, number of culverts, relationship between slope Table 2 shows example of hazard weighting for cut
and topography, distance to ridge/gully, etc. From the slopes in meta-sediments use in the SMS. The hazard
discriminant analysis, significant slope parameters that weighting was developed based on 141 cut slopes, 54 of
contributed to the landslides along the highway were it was failed slopes, in meta-sediment formations along
determined. The weightings for each parameter were the East-West Highway of Peninsular Malaysia.
then calculated using factor-overlay analysis, similar to Hazard score in percentage is computed by
the method proposed by Anbalagan[11]. The maximum summing the parameters hazard weighting of each
parameters weighting of 2 was assigned to the relatively assessed slopes and divided by the total maximum
most hazardous sub-parameters. The weighting for hazard weighting. Hazard score is then converted into
other sub-parameters is calculated using equation (1). hazard rating or hazard level as shown in Table 3.
Weighting = [Landslides frequency for sub-parameters weighting] x In 1999, the PWD developed the Slope Priority
[Maximum parameters] / [Total number of landslides] - Eq. 1 Ranking System (SPRS) as a tool for quick assessment
136
Am. J. Environ. Sci., 2 (4): 135-141, 2006

of all slopes in Malaysia so that repair work can be Table 4: Hazard score used for cut slopes used in SPRS[8]
prioritized and carried out. The SPRS is also to help Cut Slopes Hazard Score
Attributes 0 1 2
identify budget requirements for slope repairs. The i. Slope angle <450 450 - 630 >630
hazard score used in SPRS was established using very ii. Height of slope <12m 12m24m >24m
simple approach with associated ratings of 0, 1 and 2, iii. Slope cover >20% <20% -
according to the definitions of each parameter given by iv. Surface drains Good Blocked Repair
required
Hussein et al.[8]. The hazard attributes for cut slope v. Natural water path No - Yes
include slope angle, height of slope, slope cover, vi. Seepage No - Yes
surface drain, natural water path, seepage, ponding, vii. Ponding No Yes -
erosion, slope failure, surroundings upslope (human viii. Erosion Slight Moderate Critical
ix. Slope failure No - Yes
activity), soil type, weathering grade and x. Surroundings No - Yes
discontinuities. Table 4 shows hazard score use for cut upslope
slopes in the SPRS. xi. Soil type Gravel / sand Silt Clay
Hazard score in percentage is computed by
xii. Weathering grade I II, III IV- VI
summing the slope attributes hazard score of each xiii. Discontinuities No - Yes
assessed slopes and divided by the total maximum
hazard score. The hazard score is then converted into Table 5: Hazard score and rating used in the SPRS[8]
hazard rating as shown in Table 5. Cut Slope Fill Slope
In 2002, the Public Works Department (PWD) and Hazard Score Hazard Hazard Score Hazard Rating
the Japanese International Cooperation Agency (JICA) Rating
40% to 100% Very High 40% to 100% Very High
jointly developed the Slope Information Management 30% to 40% High 30% 40% High
System (SIMS)[9]. In this system, the slopes are 19% to 30% Moderate 20% 30% Moderate
assessed based on predefined likelihood of failure type 8% to 19% Low 10% 20% Low
that are base on definition used in Japan; i.e. slope 0% to 8% Very Low 0% 10% Very Low
failure/rock fall, rock mass failure, landslide, debris Y = 0.027(height) + 0.02(angle) + 0.163(shape) + 0.354(plan profile)
flow and embankment failure. The hazard score used + 0.278(cutting topography) + 0.202(structure) - 0.172(main cover
was adopted from the Japanese experience. Parameters type) + 0.472(cover) + 0.017(% rock exposure) 1.266 (corestone
considered include topography, slope geometry, slope boulders) + 0.249(rock condition profile) + 0.281(ground saturation)
4.293 - Eq.2
forming material, geological structure, any presence of
slope deformation, surface condition and
countermeasure effectiveness. Table 6 shows hazard Where Y is regression function representing
score used for slope failure/rock fall type of failure. instability score of the assessed slopes.
Table 7 shows the hazard rating applied in the SIMS. For calculation of Y, the slope parameters in the
The Slope Management and Risk Tracking bracket should be replaced by value or classes of slope
Systems (SMART) is the latest slope management variables as listed in Table 8. The equations used to
system developed by the Public Works Department[10]. transform the data from individual discriminant
The system is developed base on data from the function scores (Y) to probabilities of group
Tamparuli - Sandakan road in Sabah, East Malaysia, membership (i.e. failed or not failed) were derived
where there have been numerous slope failures. In through curve fitting. An example is shown in Table 9.
developing SMART, data from 918 cut slopes The probabilities are then grouped into groups of
comprising of 741 not failed slopes and 177 failed qualitative terms of instability category for the purpose
slopes are used. This road was underlain mainly by of interpretation and action. The instability or hazard
sediment and meta-sediment formations of mudstone, rating categories designated for this purpose are Very
sandstone and siltstone, inter-bedding each other[10]. Low, Low, Medium, High and Very High (Table 10).
The system uses slope inventory forms similar to
the SMS with some slight modifications. In SMART, FIELD STUDY SITES, SLOPES AND
the hazard score or instability score (IS) ranges from 0 LANDSLIDES INVENTORIES
to 1 and is derived through the integration of results
from three assessment methods, that is the statistical
Road is the main type of transportation system in
method (stepwise discriminant function analysis
converted into probability), deterministic method Malaysia. About 30% of these roads traversed through
(factor of safety determine by Combined Hydrology or located in hilly and mountainous areas. These
and Stability Model or CHASM and then converted to mountainous roads experience numerous landslides,
probability using Monte-Carlo simulation) and if when which cause disruption, injuries and losses to life and
appropriate, expert knowledge[10]. An example of a economy.
twelve-parameter regression equation (Equation 2) Slope inventory data from 139 cut slopes in
derived from stepwise discriminant function analysis, granitic formations along three different sites, namely
then converted into probability (P), is given as:
137
Am. J. Environ. Sci., 2 (4): 135-141, 2006

Table 6: Hazard score assign for slope failure/rock fall type of failure used in the SIMS[9]
Condition of Slope (for slope failure/rock fall) Score
Topography Alluvium slope, Trace of slope failure, Clear knick point or overhanging & Concave 0 to 2
slope or debris slope.
Geometry; select higher point of A: Soil slope H > 30m, H<30m & I>450 , 10 to 30
A or B H: High of soil 15m<H<30m & I<450 and H<15m
I: Slope angle

B: Rock slope H>50m, 30m<H<50m, 15m<H<30m 10 to 30


H: High of rock and H<30m

Material; select A and B A: Soil character; Swelling clay contents: Conspicuous, Slightly and None. 0 to 8

B: Rock quality; Sheared rock, Weathered rock: Conspicuous, Slightly and Not 0 to 8
Available.

Geological Structure Daylight structure (Planar, wedge), Soft soil over base rock, Hard rock over weak rock 0 to 8
and Others.

Deformation Slope Deformation: Erosion (gully, rill, sheet, fretting), Visible, Obscure & 0 to 10
rock fall, exfoliation etc. None

Deformation at adjacent slope (rock fall, slope failure, Visible, Obscure & 0 to 6
crack, etc.) None
Surface Condition Condition of Surface; Unstable, Moderate & Stable 0 to 8
Ground Water; Natural spring, Water seepage & Dry. 0 to 6

Cover; Bare, Grass + Structure & Structure. 1 to 4

Surface Drainage; Available (good), Available (need repair) & Not available. 0 to 2

Countermeasure effectiveness Effective, Partially effective & Not effective or No countermeasure. - 20 to 0

Total Score

Table 7: Hazard rating applied in the SIMS[9]


Level of Slope Management Hazard Score (%) Hazard Rating
Level I R>75 Very High
Level II 75>R>65 High
Level III 65>R>50 Moderate
Level IV R<50 Low

Table 8: Variables / Parameters for cut slope determined significant in SMART[10]


No. Slope Variable Range of Classes Value / Classes
1 Height Any value from 0 to 200 meters 0 to 200
2 Slope angle Any value from 0 to 90 degrees 0 to 90
3 Slope shape Simple, Planar, Asymmetrical & Compound. 1 to 4
4 Plan profile Convex, Concave & Straight 1 to 3
5 Cutting topography Top, Middle, Base, Basin/Flat Ground & Sidelong Embankment 1 to 5
6 Structure None, Crib Wall, Piled Wall, Surface Netting, Soil Nailing, 1 to 10
Gabion Wall, Rock Bolts / Stitching, Concrete Wall, Masonary
Wall & Others.
7 Main cover type Grass, Shrub, Fern, Jungle, Plantation, Agricultural & Others. 1 to 7
8 Slope cover Good (100%), Average (80 to 100%) & Poor (< 80%). 1 to 3
9 Percentage rock exposure Any number from 0 to 100 % 0 to 100
10 Corestone boulders No & Yes 0 & -1
11 Rock condition profile Majority < Grade III, Partly < Grade III & Partly > Grade IV, 1 to 5
Predominantly Grade IV to Grade VI, Predominantly Grade IV to
Grade VI but with Corestone Boulders & Predominantly
Colluvium.
12 Measure of ground saturation Low, Medium, High & Very High 0 to 3

138
Am. J. Environ. Sci., 2 (4): 135-141, 2006

Table 9: Conversion of Y into probability, P[10] the granitic formations. Whilst data from 47 cut slopes
Value of Y Calculation of probability, P in meta-sediment formation along the Gunung Raya
Y < -2 P = 0.05
-2 < Y < 0.5 P = 0.0037Y3 + 0.0891Y2 + 0.3195Y 0.3531 road and the East-West highway were used for
0.5 < Y < 4 P = 0.0105Y3 0.1275Y2 + 0.5152Y + 0.2952 evaluating the SAS in the meta-sediment formation.
Y>4 P=1 The slope inventory data such as slope height,
slope angle, soil type, weathering grade, were
Table 10: Probability and instability category use in SMART[10] collected/compiled over a ten-year period, from 1994 to
Probability, P Instability Category
2004. These data were obtained from previous record as
0.0 0.2 Very Low
0.2 0.4 Low well as through site visits (walkthrough survey).
0.4 - 0.6 Medium Landslide occurrences used were those that had
0.6 0.8 High occurred after the initial slope inventory data was
0.8 1.0 Very High collected. They were determined from written historical
records, differences seen on multi-date aerial photos, or
the Gunung Raya road in Langkawi Island, the East- difference between older sketches of the data collection
West Highway, Perak and the Kuala Kubu Baru Gap performa with the current site conditions. Table 11 and
road, Selangor, Malaysia, were used in the evaluation 12 summarize informations on the 186 numbers of cut
of the slope assessment systems (SAS) of cut slopes in slopes considered in this study.

Table 11: Cut slopes in granitic formation


Location No. of cut slopes No. of slope Date of initial Date of slope General remarks on type of slope
considered in the failures data failures failures, reasons of failure
study
Gunung Raya 34 10 April 1996 Between April Mostly shallow slides except one
road, Langkawi 1996 to November deep seated slide at KM 5.9
Island 2003
East-West 53 12 March 1996 Between March Mostly shallow slides
Highway, Perak 1996 to July 2001
Kuala Kubu Gap 52 22 August 2000 Between August Mostly shallow slides except 2
road, Selangor 2000 to November debris flow at KM 23.44 and
2003 adjacent to it

Table 12: Cut slopes in meta-sediment formation


Location No. of cut slopes No. of slope Date of initial Date of slope General remarks on type of
considered in the failures data failures slope failures, reasons of
study failure
Gunung Raya road, 12 5 April 1996 Between April Mostly shallow slides
Langkawi Island 1996 to November
2003
East-West 35 24 March 1996 Between March Mostly shallow slides
Highway, Perak 1996 to July 2001

Table 13: Accuracy of the slope assessment systems in predicting landslides


(i) Cut slopes in granitic formations
Prediction SMS SPRS SIMS SMART
(1) Number of slopes assessed 139 139 139 139
(2) Number of recent landslides or failed slopes 44 44 44 44
(3) Number of slopes classified as High and Very High Hazard 17 23 1 27
that actually failed
(4) Percentage of (3) compared with (2) 39% 52% 2% 61%

(ii) Cut slopes in meta-sediment formations


Prediction SMS SPRS SIMS SMART
(1) Number of slopes assessed 47 47 47 47
(2) Number of actual landslides or failed slopes 29 29 29 29

(3) Number of slopes classified as High and Very High Hazard 13 17 5 26


that actually failed
(4) Percentage of (3) compared with (2) 45% 59% 17% 90%

Note: SMS - Slope Maintenance System (SMS), SPRS - Slope Priority Ranking System, SIMS- Slope Information Management System
SMART - Slope Management and Risk Tracking System.

139
Am. J. Environ. Sci., 2 (4): 135-141, 2006

HAZARD ASSESSMENT OF THE SLOPE AND


ACCURACY EVALUATION OF THE SAS Table 14: Accuracy of the models from previous works by other
researches on landslide assessment
No. Country Accuracy (%) References
Accuracy or reliability in predicting future
landslide is a crucial part of any slope assessment 1 Italy 72.7 & 80.7 Carrara et al. [12]
systems (SAS). In this study, the accuracy of the SAS 2 Italy 72.0 Guzzetti et al.[13]
was determined by comparing the hazard rating of each 3 Bolivia 78 to 89 Ploquin and Gwyn[14]
of the slopes evaluated based on the initial (earlier)
slope inventory data with the later set of data, i.e. after
CONCLUSION
the landslide occurrences in some cases. The accuracy
in percentage was determined by comparing the number
From the result of this study, it is found that none
of slopes classified as high and very high hazard that
of the four slope assessment systems, namely the Slope
actually failed with the total number of actually failed
Maintenance System (SMS), the Slope Priority Ranking
slopes.
System (SPRS), the Slope Information Management
Prior to the hazard assessment exercise, the format System (SIMS) and the Slope Management and Risk
of the available data was transferred according to the Tracking System (SMART), was satisfactory in
range or classes of all SAS. Some data which was not predicting landslide in cut slopes in granitic formations,
available especially the permanent parameters related to base on slope inventory data from 139 cut slopes. The
the geometry and geological features of the assessed reasons for this range from the use of hazard score
slope were determined through site visits (walkthrough developed from another country, to insufficient
survey). Some estimates were made for the parameter database, to the use of an oversimplified approach and
value needed in each SAS such as strength parameters to the use of database derived from different rock/soil
of soil and rock, soil depth, permeability etc. formation.
Table 13 summarizes the prediction accuracy of the However for the case of cut slope in meta-
five SAS considered in the study, for cut slopes in both sediments, the Slope Management and Risk Tracking
granitic and meta-sediment formations.As shown in System (SMART) is found to be satisfactory with 90%
Table 13, none of the existing slope assessment systems prediction accuracy. The current database of the
(SAS) appeared to be satisfactory in predicting SMART is based on meta-sediment formation.
landslides in cut slopes in granitic formations.
Satisfactory in this case is defined as percentage of REFERENCES
accuracy of greater than 70% (Table 14). The reasons
for this could perhaps be explained as follows. 1. Othman, M.A. and D.M. Lloyd, 2001. Slope
For the case of the SMS (Slope Maintenance instability problems of roads in mountainous
System), it appeared that the development of SMS terrain: a geotechnical perspective. Proc. Natl.
using 74 cut slopes database that was limited to one Slope Seminar, Cameron Highland, Malaysia, pp:
site, that is the East-West Highway, was not sufficient. 11.
For the case of the SPRS (Slope Priority Ranking
2. Little, A.L., 1969. The engineering classification of
System), it uses a too simplified approach of assigning
residual tropical soils. Proc. Speciality Session on
hazard score with only 0, 1 and 2. For the case of SIMS
(Slope Information Management Systems), it uses the Engineering Properties of Lateritic Soil, Vol. 1,
hazard score developed from other country (Japan), 7th Intl. Conf. Soil Mechanics & Foundation
which appears to be its main weakness. For the case of Engineering, Mexico City, pp: 1-10.
the SMART (slope management and risk tracking 3. Varnes, D.J., 1978. Landslide types and processes.
systems), its current database derived mainly from the In Special Report 29: Landslides and Engineering
meta-sediment formations is apparently not suitable to Practice (E.B. Eckel, ed.), HRB, National Research
be extrapolated to cut slopes in other rock/soil Council, Washington, D.C., pp: 20-47.
formations. 4. Hussein, A.N., H. Omar and S. Jamaludin, 2004.
However, for case of cut slope in meta-sediment Slope assessment and management. In Huat, BBK,
formation, SMART appears to be satisfactory with a Gue, S.S. and F.H. Ali, (Eds); Tropical Residual
prediction accuracy of 90%, but not the other four SAS,
Soils Engineering, Balkema, London, pp: 103-120.
namely the SMS, SPRS and SIMS. This is perhaps not
5. Tangestani, M.H., 2003. Landslide susceptibility
so surprising for SMART as its current database is
derived mainly from the meta-sediment formations. mapping using the fuzzy gamma operation in a
This seems to reinforce the earlier argument that slope GIS, Kakan catchment area, Iran. Proc. Map of
assessment system develops for one rock/soil formation India Conf. GIS evelopment.net, pp: 7.
cannot be extrapolated to other rock/soil formation.
140
Am. J. Environ. Sci., 2 (4): 135-141, 2006

6. Yi, S., R. Li, X. Pu and S. Fu, 2000. The fractal 12. Carrara, A., M. Cardinali, F. Guzzetti and P.
characteristics of the temporal and spatial Reichenbach, 1995. GIS-Based techniques for
distribution of the Zameila Mountain landslide mapping landslide hazard. In Geographical
activities and its fracture structure in Tibet of Information System in Assessing Natural Hazards.
China. In Landslides in research, theory and Academic Publication, Dordrecht, Netherlands.
practice. Thomas Telford, London. Pp: 1605-1608. 13. Guzzetti, F., A. Carrara, M. Cardinali and P.
7. PWD Malaysia, 1996. Final Hazard Analysis Reichenbach, 1999. Landslide hazard evaluation: A
Report. In East-West Highway Long Term review of current techniques and their application
Preventive Measures and Stability Study, pp: 33. in a multi-scale study, Central Italy. J.
8. Hussein, A.N., K. Kassim, K.O. Lai and S. Geomorphol., 31: 181216.
Jamaludin, 1999. A slope management system for 14. Peloquin, S. and Q.H.J. Gwyn, 2000. Using remote
prioritisation of cut and fill slopes in Malaysia.
sensing, GIS and artificial intelligence to evaluate
10th REAAA conference: Road development for
landslide susceptibility levels: Application in the
21st Century, Tokyo, Japan, pp: 9.
9. JICA and PWD, 2002. The study on slope disaster Bolivian Andes. Proc. of 4th Intl. Conf. on
management for federal roads in Malaysia. Final Integrating GIS and Environmental Modeling
report. Vol. 1. unpublished. (GIS/EM4): Problems, Prospects and Research
10. PWD Malaysia, 2004. Slope protection study for Needs. Banff, Alberta, Canada, pp: 12.
Federal Route 22, TamparuliSandakan, Sabah.
Draft final report.
11. Anbalagan, R., 1995. Terrain evaluation and
landslide hazard zonation for environmental
regeneration and land use planning in mountainous
terrain. Proc. Sixth Symposium on Landslides,
Christchurch, New Zealand, pp: 861-868.

141

View publication stats

You might also like